Merge pull request #14 from Yolean/statefulset

Convert PetSet to StatefulSet for k8s 1.5+

Author: solsson

bootstrap/pvc.yml renamed to 10pvc.yml

@@ -9,11 +9,11 @@ spec:
     - ReadWriteOnce
   resources:
     requests:
-      storage: 100Mi
+      storage: 200Gi
   selector:
     matchLabels:
       app: kafka
-      petindex: "0"
+      podindex: "0"
 ---
 kind: PersistentVolumeClaim
 apiVersion: v1
@@ -25,11 +25,11 @@ spec:
     - ReadWriteOnce
   resources:
     requests:
-      storage: 100Mi
+      storage: 200Gi
   selector:
     matchLabels:
       app: kafka
-      petindex: "1"
+      podindex: "1"
 ---
 kind: PersistentVolumeClaim
 apiVersion: v1
@@ -41,8 +41,8 @@ spec:
     - ReadWriteOnce
   resources:
     requests:
-      storage: 100Mi
+      storage: 200Gi
   selector:
     matchLabels:
       app: kafka
-      petindex: "2"
+      podindex: "2"

20dns.yml

@@ -3,8 +3,6 @@
 apiVersion: v1
 kind: Service
 metadata:
-  annotations:
-    service.alpha.kubernetes.io/tolerate-unready-endpoints: "true"
   name: broker
   namespace: kafka
 spec:

50kafka.yml

@@ -1,5 +1,5 @@
-apiVersion: apps/v1alpha1
-kind: PetSet
+apiVersion: apps/v1beta1
+kind: StatefulSet
 metadata:
   name: kafka
   namespace: kafka
@@ -10,14 +10,11 @@ spec:
     metadata:
       labels:
         app: kafka
-      annotations:
-        pod.alpha.kubernetes.io/initialized: "true"
-        pod.alpha.kubernetes.io/init-containers: '[
-        ]'
     spec:
+      terminationGracePeriodSeconds: 10
       containers:
       - name: broker
-        image: solsson/kafka-persistent:0.10.1.0
+        image: solsson/kafka-persistent:0.10.1@sha256:110f9e866acd4fb9e059b45884c34a210b2f40d6e2f8afe98ded616f43b599f9
         ports:
         - containerPort: 9092
         command:
@@ -30,11 +27,8 @@ spec:
   volumeClaimTemplates:
   - metadata:
       name: datadir
-      namespace: kafka
-      annotations:
-        volume.alpha.kubernetes.io/storage-class: anything
     spec:
       accessModes: [ "ReadWriteOnce" ]
       resources:
         requests:
-          storage: 100Mi
+          storage: 200Gi

README.md

@@ -1,23 +1,19 @@
 
 # Kafka as Kubernetes PetSet
 
-Example of three Kafka brokers depending on three Zookeeper instances.
+Example of three Kafka brokers depending on five Zookeeper instances.
 
-To get consistent service DNS names `kafka-N.broker.kafka`(`.svc.cluster.local`), run everything in a namespace:
+To get consistent service DNS names `kafka-N.broker.kafka`(`.svc.cluster.local`), run everything in a [namespace](http://kubernetes.io/docs/admin/namespaces/walkthrough/):
 ```
 kubectl create -f 00namespace.yml
 ```
 
 ## Set up volume claims
 
-This step can be skipped in clusters that support automatic volume provisioning, such as GKE.
+You may add [storage class](http://kubernetes.io/docs/user-guide/persistent-volumes/#storageclasses)
+to the kafka StatefulSet declaration to enable automatic volume provisioning.
 
-You need this step in Minikube.
-
-```
-./zookeeper/bootstrap/pv.sh
-kubectl create -f ./zookeeper/bootstrap/pvc.yml
-```
+Alternatively create [PV](http://kubernetes.io/docs/user-guide/persistent-volumes/#persistent-volumes)s and [PVC](http://kubernetes.io/docs/user-guide/persistent-volumes/#persistentvolumeclaims)s manually. For example in Minikube.
 
 ```
 ./bootstrap/pv.sh
@@ -26,21 +22,27 @@ kubectl create -f ./bootstrap/pvc.yml
 kubectl get pvc
 ```
 
-The volume size in the example is very small. The numbers don't really matter as long as they match. Minimal size on GKE is 1 GB.
-
 ## Set up Zookeeper
 
-This module contains a copy of `pets/zookeeper/` from https://github.com/kubernetes/contrib.
+There is a Zookeeper+StatefulSet [blog post](http://blog.kubernetes.io/2016/12/statefulset-run-scale-stateful-applications-in-kubernetes.html) and [example](https://github.com/kubernetes/contrib/tree/master/statefulsets/zookeeper),
+but it appears tuned for workloads heavier than Kafka topic metadata.
 
-See the `./zookeeper` folder and follow the README there.
+The Kafka book (Definitive Guide, O'Reilly 2016) recommends that Kafka has its own Zookeeper cluster,
+so we use the [official docker image](https://hub.docker.com/_/zookeeper/)
+but with a [startup script change to guess node id from hostname](https://github.com/solsson/zookeeper-docker/commit/df9474f858ad548be8a365cb000a4dd2d2e3a217).
 
-An additional service has been added here, create using:
+Zookeeper runs as a [Deployment](http://kubernetes.io/docs/user-guide/deployments/) without persistent storage:
 ```
-kubectl create -f ./zookeeper/service.yml
+kubectl create -f ./zookeeper/
 ```
 
+If you lose your zookeeper cluster, kafka will be unaware that persisted topics exist.
+The data is still there, but you need to re-create topics.
+
 ## Start Kafka
 
+Assuming you have your PVCs `Bound`, or enabled automatic provisioning (see above), go ahead and:
+
 ```
 kubectl create -f ./
 ```

bootstrap/pv-template.yml

@@ -5,7 +5,7 @@ metadata:
   name: datadir-kafka-0
   labels:
     app: kafka
-    petindex: "0"
+    podindex: "0"
 spec:
   accessModes:
   - ReadWriteOnce
@@ -20,7 +20,7 @@ metadata:
   name: datadir-kafka-1
   labels:
     app: kafka
-    petindex: "1"
+    podindex: "1"
 spec:
   accessModes:
   - ReadWriteOnce
@@ -35,7 +35,7 @@ metadata:
   name: datadir-kafka-2
   labels:
     app: kafka
-    petindex: "2"
+    podindex: "2"
 spec:
   accessModes:
   - ReadWriteOnce

zookeeper/00namespace.yml

@@ -0,0 +1,5 @@
+---
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: kafka

zookeeper/20zoo-service.yml

@@ -0,0 +1,15 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: zoo
+  labels:
+    app: zookeeper
+spec:
+  ports:
+  - port: 2888
+    name: peer
+  - port: 3888
+    name: leader-election
+  clusterIP: None
+  selector:
+    app: zookeeper

zookeeper/service.yml renamed to zookeeper/30service.yml

@@ -9,4 +9,4 @@ spec:
   - port: 2181
     name: client
   selector:
-    app: zk
+    app: zookeeper

zookeeper/50zoo.yml

@@ -0,0 +1,38 @@
+apiVersion: apps/v1beta1
+kind: StatefulSet
+metadata:
+  name: zoo
+  namespace: kafka
+spec:
+  serviceName: "zoo"
+  replicas: 5
+  template:
+    metadata:
+      labels:
+        app: zookeeper
+    spec:
+      terminationGracePeriodSeconds: 10
+      containers:
+        - name: zookeeper
+          image: solsson/zookeeper-statefulset:3.4.9@sha256:d32b44b32009a69b3450a5216f459e504f1041f587596895219fc04cf22f5546
+          env:
+            - name: ZOO_SERVERS
+              value: server.1=zoo-0.zoo:2888:3888:participant server.2=zoo-1.zoo:2888:3888:participant server.3=zoo-2.zoo:2888:3888:participant server.4=zoo-3.zoo:2888:3888:participant server.5=zoo-4.zoo:2888:3888:participant
+          ports:
+            - containerPort: 2181
+              name: client
+            - containerPort: 2888
+              name: peer
+            - containerPort: 3888
+              name: leader-election
+          volumeMounts:
+            - name: datadir
+              mountPath: /data
+            # There's defaults in this folder, such as logging config
+            #- name: conf
+            #  mountPath: /conf
+      volumes:
+        #- name: conf
+        #  emptyDir: {}
+        - name: datadir
+          emptyDir: {}